ChangeSignal
Enables querying and analyzing changes in technology company documentation through MCP tools, including change feeds, company info, and evidence-based messaging.
README
ChangeSignal
ChangeSignal is a source-backed product-intelligence shell for tracking how technology companies change their public documentation over time.
It is designed to answer:
- What did a company ship or change?
- Which product areas are receiving the most investment?
- How quickly is each company moving?
- Which meaningful changes never appeared in the public changelog?
- What evidence can GTM, customer-success, support, and competitive-intelligence teams use in messaging?
This repository does not contain a crawler. It provides the web product, normalized ingestion contract, durable data boundaries, CLI, HTTP/OpenAPI, and MCP surfaces required to integrate an existing crawler later.
Current status
The repository is a usable, fake-safe product shell with:
- overview, change feed, evidence detail, companies, product areas, subscriptions, analytics, documentation sources, and messaging views;
- five synthetic companies and fifteen synthetic documentation changes;
- seventeen shared operations projected across web, HTTP, CLI, and MCP;
- a real MCP stdio transport;
- a versioned
crawler-change-batch.v1validation boundary; and - explicit preview or
not-configuredresponses for actions that would require persistence, delivery, crawling, or a model provider.
Fixture data is synthetic and does not represent live crawler output.
Run the web app
Requirements: Node.js 22 and pnpm 10.12.1.
pnpm install --frozen-lockfile
pnpm --dir apps/web dev --port 8421 --strictPort
Open http://127.0.0.1:8421.
Primary routes include:
/— product-intelligence overview/changes— searchable change feed/companies— monitored-company catalog/subscriptions— watchlists and digest preferences/analytics— velocity, investment areas, and changelog gaps/sources— documentation-source readiness/agents— evidence-grounded messaging analyst/api— API, CLI, and MCP catalog
CLI
pnpm changesignal overview
pnpm changesignal companies list
pnpm changesignal changes list --company relaydesk
pnpm changesignal changes get relaydesk-routing-controls
pnpm changesignal sources list --company relaydesk
pnpm changesignal messaging brief relaydesk-routing-controls
pnpm changesignal ingestion validate \
--file docs/examples/change-signal/crawler-change-batch.v1.json
pnpm changesignal mcp tools
Add --json to supported commands for the shared machine-readable operation
envelope.
MCP
Launch the stdio server with:
pnpm changesignal mcp serve
The server supports MCP initialization, tool discovery, reviewed read tools, source-mapping previews, messaging briefs, and bounded crawler-batch validation. Persistence mutations remain disabled until authentication, audit, idempotency, and storage are connected.
Integrating crawler output
The crawler owner supplies one redacted representative payload plus transport,
authentication, stable-ID, retry, replay, and snapshot-reference details. A
provider adapter then maps the native payload into crawler-change-batch.v1;
crawler-specific logic stays outside the web, CLI, MCP, capability, and workflow
layers.
See:
- Product plan
- Crawler integration handoff
- Example normalized batch
- Release evidence
- Factory evaluation and papercuts
Verification
Focused product verification:
pnpm --dir packages/change-signal test
pnpm --dir packages/change-signal typecheck
pnpm --dir apps/cli test
pnpm --dir apps/cli typecheck
pnpm --dir apps/web test
pnpm --dir apps/web typecheck
pnpm --dir apps/web build
The generated foundation retains several upstream factory verification defects.
They are documented without weakening the affected tests in
docs/factory-evaluation.md.
Architecture
web routes -> screens -> features -> blocks -> Saas UI/shared primitives
API/CLI/MCP -> shared operation registry -> capabilities/workflows
crawler provider adapter -> normalized batch -> validation -> durable owners
storage/notifications/observability -> provider adapters
The application was generated from the Maestro SaaS UI factory and then built as a product-specific shell. The factory evaluation is retained because this repository also serves as a real-world test of that application factory.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
E2B
Using MCP to run code via e2b.